4

I have this python code for displaying some numbers over time:

import matplotlib.pyplot as plt
import datetime
import numpy as np

x = np.array([datetime.datetime(2013, 9, i).strftime("%Y-%m-%d") for i in range(1,5)], 
            dtype='datetime64')
y = np.array([1,-1,7,-3])
plt.plot(x,y)
plt.axhline(linewidth=4, color='r')
plt.show()

The resulting graph has the numbers 0.0 to 3.0 on the x-axis:

enter image description here

What is the simplest way to display dates instead of these numbers? Preferably in the format %b %d.

unutbu
  • 842,883
  • 184
  • 1,785
  • 1,677
vfxGer
  • 311
  • 3
  • 12

1 Answers1

5

According to efiring, matplotlib does not support NumPy datetime64 objects (at least not yet). Therefore, convert x to Python datetime.datetime objects:

x = x.astype(DT.datetime)

Next, you can specify the x-axis tick mark formatter like this:

xfmt = mdates.DateFormatter('%b %d')
ax.xaxis.set_major_formatter(xfmt)

import matplotlib.pyplot as plt
import matplotlib.dates as mdates
import datetime as DT
import numpy as np

x = np.array([DT.datetime(2013, 9, i).strftime("%Y-%m-%d") for i in range(1,5)], 
            dtype='datetime64')
x = x.astype(DT.datetime)
y = np.array([1,-1,7,-3])
fig, ax = plt.subplots()
ax.plot(x, y)
ax.axhline(linewidth=4, color='r')
xfmt = mdates.DateFormatter('%b %d')
ax.xaxis.set_major_formatter(xfmt)
plt.show()

enter image description here

unutbu
  • 842,883
  • 184
  • 1,785
  • 1,677
  • Actually I only used the NumPy datetime64 object because I thought it would help so just x = np.array([datetime.datetime(2013, 9, i) for i in range(1,5)]) does it. – vfxGer Feb 26 '14 at 20:42